Effect of Pioglitazone Administered to Patients With Adrenomyeloneuropathy

Proof of concept for this trial is provided by the results of biochemical, neuropathological and motor effects of pioglitazone in two mouse models of AMN. Pioglitazone was given orally (9 mg/kg/day) for two months in both models.

The Abcd1‐null mouse model already shows at 3,5 months biochemical signs oxidative stress that increase with time and are then associated with energy homeostasis alterations, although first clinical signs of AMN—i.e. axonopathy and locomotor impairment—appear at 20 months. In these mice, there are mitochondrial anomalies, decreased levels of PGC‐1α which is a master regulator of mitochondrial biogenesis, and decreased levels and activity of SIRT1α, which activates PGC‐1α.

The Abcd1‐null mouse can be considered as a "AMN‐like" model, because of the absence of demyelinating lesions in brain and spinal cord, the presence of non‐inflammatory ''dying‐back'' axonopathy in peripheral nerves and spinal cord and its late‐onset motor deficits that all are hallmarks of AMN in X‐ALD patients. This model was used to assess the efficacy of pioglitazone on several biochemical markers in the spinal cord of Abcd1‐null mice (N=12), using comparisons with placebo-¬treated Abcd1‐null mice (N=12) or wild‐type mice (N=12).

In Abcd1‐null mice treated with pioglitazone at 10,5 months of age and studied at 12 months (1,5 months following the beginning of the ongoing treatment), mitochondrial anomalies were corrected to the level of wild type control mice. Indeed, mitochondrial DNA and protein (including PGC‐1α, NRF1 and TFAM) levels were corrected; as well as mitochondrial metabolism, as assessed by pyruvate kinase activity, ATP and NAD+ concentrations. Pioglitazone had no effect on SIRT1 expression (mRNA and protein levels). However, pioglitazone significantly lowered the carbonylation of SIRT1 protein, which presumably accounts for the observed rescue of SIRT1 activity.

In these mice treated with pioglitazone, oxidative lesions in the spinal cord were reversed. Studied oxidative stress biomarkers included markers of oxidative lesions to proteins (GSA, AASA), lipids (MDAL) and carbohydrates (CEL). Additionally, the activity and concentration level of antioxidant enzymes GPX1, which were increased in untreated Abcd1‐null mice, but not SOD2, was normalized to the level of wild type mice.

The second mouse model is the double knockouts (DKO) in which both Abcd1 and Abcd2 transporters are inactivated. The Abcd1‐/Abcd2‐/‐DKO exhibits greater VLCFA accumulation in spinal cord (Pujol et al., 2004), higher levels of oxidative damage to proteins, and a more severe AMN-¬like pathology, with earlier onset of motor impairment than the single Abcd1‐null mouse (12 months in the DKO compared to 20 months in Abcd1‐null mice). Efficacy of pioglitazone at the motor and neuropathologic levels was studied in 17 Abcd1‐/Abcd2‐/‐mice comparing them with placebo‐treated Abcd1‐/Abcd2-/-mice (N=17) and wild‐type mice (N=25).

In Abcd1‐/Abcd2‐/‐mice treated with pioglitazone at 13 months of age and studied at 15 or 17 months (treatment duration of 2 to 4 months), axonal degeneration was prevented, as shown by the normalization to the control level of number of APP or synaptophysin positive axons.

Also, pioglitazone arrested the progression of locomotor deficits in these mice, as assessed by the treadmill test and the bar‐cross test. Indeed, the locomotor performances of pioglitazone DKO after four months of treatment mice reached the performances of the controls.

Overall, these studies show the efficacy of treatment with pioglitazone in "AMN‐like mice "and provide a strong rationale for conducting a preliminary open clinical trial with pioglitazone in AMN patients.